CAN receive does not work properly on S32K148. Receives less than half of the messages.

[Poley]

Hello, this is an urgent issue so would really appreciate the help.

I have just connected my board to all of my physical nodes (17) and noticed that I was sometimes missing some data updates. I decided to plot some CAN signals to see what was happening:

The blue signal is from 1 node (Non NXP S32) that is sent to my S32k board. In my model the received signal is sent straight back out onto CAN, this is the green signal. As you can see I am missing a very large amount of data, this is constant and is never similar to the CAN data that it should be mirroring.

The second node in pink is represented more accurately in the mirrored red signal, it still misses 1.7s of data in the middle there and has random dropouts. This data is critical and we can not miss more than a few within a short period. 

This is one of my receive blocks, they are all the same layout, just with different ID'd and MB's:

Help will be greatly appreciated, thank you!

[MuhammedKoyuncu]

Hi

Anyone able to find a solution to the problem?

[mariuslucianand]

Hello @MuhammedKoyuncu,

Can you tell us please, which derivative are you using from the MPC57xx family?

Regards,

Marius

[Poley]

Hi Marius,

I don't think my masks and MB's are being set up properly in generated code?

IFLAG1 never changes and my masks do not seem to occupy the 32 MB's with individual masks:

[MuhammedKoyuncu]

Hi @mariuslucianand 

Device: RDVCU5775EVM

Regards,

Muhammed

[Poley]

Hi Muhammed,

Have you got anywhere further with this by any chance?

My chip still misses lots of messages, sometimes missed 40s worth of messages at a time which equates to 4,000 messages! I think it is a bug with the code generated in the toolbox at the point!

Thanks 

[MuhammedKoyuncu]

Hi

I am facing same issue with MPC57xx device.
It usually processes received messages with a delay.
1-) For this, I tried to do the operations with Rx FIFO. But nothing has changed.
2-) I tried to keep the RX FIFO Handler short, so I got the data and moved the decode outside. But the result is still the same.
3-) Finally, I'm thinking of changing the interrupt priority. I'm curious about the result.
The solution to this problem is really important.
I can't think of any other ideas. What could happen?

Regards,

Muhammed

[MuhammedKoyuncu]

This is the last one.Still a lot of data loss.

Regards,

Muhammed